Kamil

Founder @SmophyAI - building an AI workspace that turns multiple models into real work, not just chat.

[Kamil @madebykamil]

@CarterSailing Thanks! 😀

[Kamil @madebykamil]

Google pays SpaceX for compute. Anthropic pays SpaceX for compute. Now the Pentagon might too.

While everyone was racing to build the best model, Musk built the ground they all have to stand on.

That's not model strategy. That's infrastructure chess.

[Kamil @madebykamil]

All the latest AI models unanimously predict:

🏆 Spain 2–1 Argentina in the 2026 World Cup final.

Surprising and intriguing.

[Kamil @madebykamil]

@KittyTellesWolf

[Kamil @madebykamil]

@AmaechiJust_ @elonmusk True!

[Kamil @madebykamil]

Gemini 3.5 Pro just missed its third deadline. Google scrapped the whole base model and rebuilt from scratch, still not enough.

GPT-5.6, Grok 4.5, and DeepSeek V4 all shipped this month.

Prediction markets now say August 7. Being early isn't the hard part. Shipping is.

[Kamil @madebykamil]

Today, two things happened on opposite sides of the planet at the same time.

Google is expected to ship Gemini 3.5 Pro. In Shanghai, Xi Jinping made his first personal appearance at the World AI Conference since it started in 2018.

One company, one country. Same day.

[Kamil @madebykamil]

So, how are your ideas going today?
Pre-Grok or already post-Grok? 😏

[Kamil @madebykamil]

Gemini 3.5 Pro is reportedly launching July 17 with a 2M-token context window, double what most frontier models offer today.

If the leaks hold, Google is betting on context and distribution, not raw benchmarks. Nothing official from Google yet.

[Kamil @madebykamil]

@sama GPT-5.6 Sol feels noticeably stronger than Claude Fable 5. It analyzes big files incredibly well and gives me exactly what I need every time. When I search for anything, I get complete, detailed answers. It’s become my go-to model right now.

[Kamil @madebykamil]

Apple just sued OpenAI. The claim: over 400 former Apple employees now work there, and Apple says they carried trade secrets out the door to build OpenAI's hardware device.

Two years ago they were partners. Now the AI hardware race is a courtroom fight.

[Kamil @madebykamil]

Meta's new AI detector failed to verify 55% of its own AI images after they were simply cropped.

The tool built to catch deepfakes can't catch Meta's own work once you crop it. Detection is years behind generation.

[Kamil @madebykamil]

@finkd Ran the numbers: on SWE-Bench Pro, Muse Spark 1.1 gives 14.47 score per dollar of output tokens. Grok 4.5 gives 10.78. Opus 4.8 gives 2.77.

Full breakdown here: https://t.co/beiQkpR2zj

[Kamil @madebykamil]

Grok 4.5 was the cheapest near-frontier model from a major US lab. It held that for one day. Meta took it.

SWE-Bench Pro, score per dollar of output tokens:

Muse Spark 1.1 - 14.47
Grok 4.5 - 10.78
Opus 4.8 - 2.77
GPT-5.5 - 1.95

xAI and Meta published these tables separately. Both put Opus 4.8 at 69.2 and GPT-5.5 at 58.6, which is what makes the comparison worth running.

Meta leads tool use. Opus leads SWE-Bench Pro. GPT-5.5 leads the other coding benchmarks. Nobody wins everything.

[Kamil @madebykamil]

Grok 4.5 was the cheapest near-frontier model from a major US lab. It held that for one day. Meta took it.

SWE-Bench Pro, score per dollar of output tokens:

Muse Spark 1.1 - 14.47
Grok 4.5 - 10.78
Opus 4.8 - 2.77
GPT-5.5 - 1.95

xAI and Meta published these tables separately. Both put Opus 4.8 at 69.2 and GPT-5.5 at 58.6, which is what makes the comparison worth running.

Meta leads tool use. Opus leads SWE-Bench Pro. GPT-5.5 leads the other coding benchmarks. Nobody wins everything.

[Kamil @madebykamil]

@elonmusk Grok 4.5 goes public Wednesday. Mark ships Muse Spark on Thursday. On your platform. Three years later he finally showed up to the fight, just picked a different cage.

[Kamil @madebykamil]

Grok 4.5 is 1st on intelligence per dollar. By 4x.

9.00 vs Opus 4.8 at 2.24, GPT-5.5 at 1.83, Fable 5 at 1.20.

On SWE Bench Pro it scores 64.7% to Opus 4.8's 69.2%, but costs $0.096 per task vs $1.68. 17.5x cheaper.

Frontier intelligence stopped being a spending contest.

[Kamil @madebykamil]

@CryptoWhale He is a real robot.

[Kamil @madebykamil]

OpenAI: founded 2015, ChatGPT shipped 7 years later. Anthropic: founded 2021, Claude shipped 2 years later. xAI: founded March 2023, Grok shipped its first public beta 8 months later. Same industry. Wildly different clock speed.

[Kamil @madebykamil]

@RohanJ_Markets @SwanDesk We’ll see what it looks like in a year or two. Healthy competition is good, it pushes everyone to constant development. We’ll see where the border is.